Need for expanded Candida Score for empiric antifungal use in medically critically ill patients?

Excerpt

Candida spp. rank among the top four nosocomial bloodstream infections in critically ill patients with up to 40% mortality despite antifungal therapy [1]. The utility of biomarkers such as β-1,3-d-glucan and procalcitonin, alone or in combination, demonstrate promise; nevertheless, bedside scoring tools are useful for guiding clinical decision-making [2]. Recent guidelines recommend the use of risk prediction tools to facilitate earlier recognition and initiation of antifungal therapy [3]. Commonly cited is the “Candida Score,” which showed sensitivity and specificity for invasive candidiasis of 81% and 74%, respectively, for scores > 2.5 [4]. This was a mixed medical-surgical intensive care unit (ICU) patient population, with only 35% of admissions for medical reasons. Thus, the application of this tool for patients with nonsurgical reasons for ICU admission may be poor. Additionally, certain components of the score, surgery and parenteral nutrition (PN), may be less applicable to the medical ICU (MICU) population. Although other predictive tools have been developed, they have not been validated prospectively like the Candida Score and carry low positive predictive values. …